[Spacewalk-list] Provisioning Issue... ?

[Jan Pazdziora]

On Mon, Aug 01, 2011 at 09:07:46AM -0400, Wojtak, Greg wrote:
> I'm not really sure if this is related to Spacewalk or if I've got something else going on.  I've successfully gotten bare metal provisioning to work, and I've almost got it working the way I want.  The only outstanding issue I'm seeing is when I associate my updates channel ('updates' being the channel associated with the CentOS 6 updates channel), the install dies installing ntpdate.  It looks like it is failing because the %post script for the package is exiting with a status 127.  I also noticed that almost every other rpm package is exiting with the same status (especially libraries, presumably running ldconfig in the %post), except that they are warning out rather than erroring out.  If I install disassociate the updates channel, I don't see these errors any more and ntpdate installs just fine and a yum update after the install completes successfully.  I checked md5sums of the actual rpm's and the package is not corrupt.
> 
> I'm not sure if this is an issue with the ntpdate package or if I have something set wrong in the Kickstart profile.  My first guess would be towards the package, but it kind of bothers me that all the packages' %post scripts are exiting with a status 127 (as seen from enabling logs and debugging during the kickstart).  Anyone have any ideas with this one?
> 

No idea, except: to figure out if it's Spacewalk or CentOS issue, just
put the content to plain yum repositories, store the .ks file
somewhere and update it to point to the yum repositories, and just
kickstart that .ks. If it fails as well, without Spacewalk in the
chain, it's most likely CentOS' issue.

-- 
Jan Pazdziora
Principal Software Engineer, Satellite Engineering, Red Hat